The Survey of English Usage
Annual Report 1997
There has been steady progress on the Internet Grammar of English (IGE), which we hope to complete in the course of 1998. IGE will be freely available over the Web, but we will also release a CD-ROM version.
We hope to release ICE-GB on CD-ROM in May.
The Survey is pleased to announce new funding by the Economic and Social Science Research Council (ESRC) to develop the ICECUP search facility for ICE-GB with Fuzzy Tree Fragments (FTFs).
The EPSRC Survey Parser project has now been successfully completed.
See below for details of all the above projects.
This year saw the launch of English Language and Linguistics (ELL), a new journal published by Cambridge University Press. ELL is edited by Bas Aarts in the English department at UCL and by David Denison and Richard Hogg in the Department of English and American Studies at the University of Manchester. It is published twice a year in May and November. This new international journal focuses on the description of the English language within the framework of contemporary linguistics. English Language and Linguistics will carry articles and short discussion papers or squibs on all core aspects of English, from its beginnings to the present day, including syntax, morphology, phonology, semantics, pragmatics, corpus linguistics and lexis. Submissions are welcome and can be sent to UCL. You can visit the journal's website: ELL Journal.
For the most up-to-date information about the Survey of English Usage, including the availability of forthcoming software, please check the rest of our website.
New Project: Fuzzy Tree Fragments
The Survey was awarded funding by the ESRC to extend the ICECUP search facility with Fuzzy Tree Fragments (FTFs). These allow users to specify fragments of tree structures to search the corpus for matching portions of trees. FTFs may be combined with logic to express more complex queries. We hope to make a sophisticated version of this system available, which will allow linguists to specify a search by drawing (part of) a tree on their screen, rather than having to write an expression in logic. Preliminary results show that this method will be easy to use, and impressively fast to process, due to the fact that each individual element of a tree is separately indexed. (Simple single-element searches are very fast: under 2 seconds on a 100MHz Pentium from the hard disk), while compound element (FTF) searches are better than proportional to the number of candidates. Obviously, actual performance will vary according to the computer being used and the speed of the drive on which the corpus resides. We will provide a variety of installation options to allow users to move the entire corpus, or portions, onto their hard disk. Other facilities will include a capacity to perform concordances on lexical items and entire linguistic structures, and simple statistical facilities. Updates to the software will be possible via the Survey web page. The ESRC funding was awarded to extend, improve and evaluate the software.
New Project: Landmarks in English Grammar
Landmarks in English Grammar: The Eighteenth Century is a collection of five classic eighteenth-century grammars of English. They are bundled on a single CDROM together with a copy of Acrobat Reader, the software used to view the texts. The texts in this collection have been chosen for their importance in the history of English grammar. The collection consists of the following grammars:
Charles Gildon & John Brightland, A Grammar of the English Tongue (1711). Gildon & Brightland's Grammar shows the continued influence into the eighteenth century of the PortRoyal Grammaire (1660).
Joseph Priestley, Rudiments of English Grammar (1761). Priestley's Rudiments is notable for the weight it gives to usage and custom over grammatical rules. It was used as a school textbook, and was reprinted eight times up to 1800.
Robert Lowth, A Short Introduction to English Grammar (1762). Lowth's Short Introduction is one of the bestknown English grammars. It is a central text in the prescriptive tradition, and was very influential in the work of later grammarians.
John Ash, Grammatical Institutes (1763). Ash's Grammatical Institutes was originally published in 1760. In Landmarks, we have reproduced the fourth edition, in which Ash linked his work directly to Lowth's, under a new subtitle, An Easy Introduction to Dr Lowth's English Grammar.
Lindley Murray, English Grammar (1795). Murray's English Grammar became the most popular grammar throughout the nineteenth century. Over three hundred editions are known, in Britain and America.
Each page has been optically scanned from an original edition, usually the first. Users can read the scanned pages just as they would read the original printed version, or they can use the contents list, which contains hypertext links to each chapter or section. For more precise searching, a comprehensive index allows users to find grammatical terms, citations from writers such as Swift, Pope, Addison, and Steele, and topics of contemporary interest, such as "correctness", "usage", "English and Latin compared", and the "simplicity" of English.
Landmarks in English Grammar will be available from the Survey of English Usage early in 1998.
The International Corpus of English
At the ICAME conference in Chester it was decided that Professor Chuck Meyer (firstname.lastname@example.org) would take over the coordination of the International ICE project. ICE-GB will continue to be led from the Survey, where advice will be given to international teams by Gerry Nelson.
A CD-ROM containing the ICE-GB corpus as well as The International Corpus of English Corpus Utility Programme (ICECUP) is now almost ready, and will be released in May. See http://www.ucl.ac.uk/english-usage for the most up-to-date information.
The Survey's software for searching, browsing and analysing the corpus, ICECUP III, contains facilities to perform fast searches for individual lexical items, tags, and individual nodes of a tree. It will also include a facility to perform a search for "fuzzy tree fragments", described above.
A paper on the human-computer aspect of the Survey's work in checking the ICEGB corpus (Wallis & Nelson, 1997) was presented to the European workshop on Knowledge Acquisition in Catalonia, Spain. The ICEGB corpus checking procedure was treated as a case study to compare the Survey tree editor ICE Tree II (written by Sean Wallis) and the earlier editor, ICE Tree. The large scale of the task is unprecedented in the KA community, a perhaps unexpected spinoff of the work on ICEGB.
ICE Tree II is available from the Survey, and is free for ICE Teams upon application from the team coordinator. A time-limited demonstration version can be downloaded from the Survey Web Site.
The Internet Grammar of English (JISC/JTAP Project No. 49)
We have made good progress on the Internet Grammar. Sections on all the word classes have now been written, as well as sections on phrases and clauses. Most of these are currently available online to reviewers, to whom we give access via a login and password. We have received useful feedback from several of these reviewers. Please let us know if you would like to have access to the site. We welcome all comments.
The Internet Grammar will include a comprehensive Glossary of grammatical terms, and an Index. The Glossary is currently being compiled, as the grammar is being written.
We are working on various ways to make the Grammar more interactive. We have incorporated interactive exercises into the text pages, and we are experimenting with animated graphics, which we hope to use to illustrate various features of phrase and clause structure. We have been testing the Grammar on a range of platforms, to ensure that is it accessible to as many people as possible.
The Survey Parser Project (EPSRC Grant No. GR/K75003)
This UK EPSRC-funded project finished successfully at the end of October 1997, and the final report has been submitted. The parser exploits ICEGB to inductively generate rules governing the
phrase structure of 'canonical grammatical phrases', i.e. adverb, adjective, noun, prepositional, and verb phrases. These rules are qualitative, rather than probabilistic, which means that it is far easier to modify them by hand, and they can be more linguistically meaningful. The parser is robust, fast, and fairly accurate, achieving rates of 70% coverage of noun phrases extracted from dictionary definitions, with 90% having an exact match with an annotation performed independently by hand. Since this is a parser that has been trained on a corpus, we expect the coverage to improve as the range and quality of the original corpus annotation is improved. Four samples of about 68,000 words each were analysed for verb phrase boundaries, internal structures, and labels. Over 95% accuracy was achieved for all of the samples (Fang, 1997). To support the parser, a largescale lexical database was constructed, with 160,000 individually indexed entries, each of a distinct word form. The database includes crossreference information (word forms, derivations, compounds etc.), which is supported by semantic information specifying the classes of nouns, verbs and adjectives. In the project, this information was used to support the parser when deciding between functional types of prepositional phrases.
Celine Bijleveld has left to take up a full-time appointment. We thank her for the work she did for us and wish her luck in her new position as sub-editor at EL Gazette.
Judith Broadbent was appointed lecturer at the Roehampton Institute. We wish her luck in her new job, and thank her for her work in the Survey over the past five years.
Justin Buckley continues to work as a Web designer on the IGE project. He has also taken care of the technical implementation of the Landmarks CD-ROM, and has an advisory role on the writing of the ICE-GB manual.
Marie Gibney continues as the Survey's administrator. Her knowledge of the Survey and such matters as funding application procedures has been invaluable. We hope that she will remain with us in the Survey well into the next century.
Isaac Hallegua continues to work on systems and data management. His assistance is highly valued.
Gerry Nelson is principally employed on the IGE project, for which he writes the content. He is also working on the ICE project and initiated the Landmarks project described above. Needless to say, his knowledge of the ICE project is invaluable to us, and to ICE teams worldwide.
René Quinault has been working on re-checking some of the more difficult transcriptions in ICE- GB, as well as on the splitting of the text soundfiles in line with the sentence numbering. He has also looked after the needs of visitors requiring to study the recordings of the first Survey corpus. We are grateful for his continued support.
Sean Wallis has worked on the EPSRC parser project, and full-time on ICECUP since the parser project came to an end. He designed, and will be working on, the new ESRC funded FTF project described above.
Jonathan White works on IGE, as well as on the ICE-GB project, specifically the ICE-GB manual.
Aarts, Bas, (1997) English syntax and argumentation. Macmillan Modern Linguistics Series. Basingstoke and London: Macmillan.
Aarts, Bas (1997) The role of argumentation in the description of English. In: Jan Aarts, Inge de Mönninck and Herman Wekker (eds.) Studies in English language and teaching. Amsterdam: Rodopi.
Aarts, Bas (1997) Predicative XPs in English, Journal of English Linguistics, 25.4, 332-339.
Aarts, Bas (1997) Binominal Noun Phrases revisited. Paper presented at the Research Centre for English and Applied Linguistics, University of Cambridge.
Aarts, Bas (1997) Lecture series on syntax and argumentation. Universidad de La Laguna, Tenerife, Spain. April 4-13.
Aarts, Bas and Gerald Nelson (1997) New horizons in grammar teaching: the Internet Grammar of English. Paper presented at the eighteenth ICAME conference, Chester, UK. May 21-25.
Aarts, Bas (1997) Constructing text corpora. Paper presented at the Methodology Institute, London School of Economics and Political Science, University of London.
Cairncross, Andrew (1997) Positional variation of the adjunct only in written British English. Journal of English Linguistics 25.1. 59-75.
Declerck, Renaat, (1997) The past perfect with future time reference. English Language and Linguistics 1.1, 49-61
Elsness, Johan, (1997) The perfect and the preterite in contemporary and earlier English. Topics in English Linguistics 21. Berlin and New York: Mouton de Gruyter.
Fang, Alex Chengyu (1997) Verb Forms and Subcategorisations. Oxford Literary and Linguistic Computing, 12.4. 209217.
Fang, Alex Chengyu and S. Yamazaki (1997) The International Corpus of English and TEFL: In Memory of Professor Sidney Greenbaum. Daito Gogaku Kyoiku Ronshu 5. 739.
Fang, Alex Chengyu (1997) Prepositional Phrases: Towards the Automatic Determination of their Syntactic Functions. Paper presented at the18th ICAME Conference, Chester, UK. May 21-25.
Hoye, Leo (1997) Adverbs and Modality in English. London: Longman.
Ljung, Magnus, (1997) A genre-based study of English subordinator-headed non-finite and verbless adverbial clauses. In: To explain the present: Studies in the changing English Language in honour of Matti Rissanen. Terttu Nevalainen and Leena Kahlas-Tarkka (eds.) Helsinki: Societé Néophilologique. Mémoires de la Société Néophilologique de Helsinki, LII, 375-394
Nelson, G. (1997) A study of the top 100 wordforms in the ICE-GB text categories. International Journal of Lexicography, 10.2, 112-134.
Nelson, G. (1997) Standardising wordforms in a spoken corpus. Literary and Linguistic Computing. 12.2, 79-85.
Nelson, G. (1997) Review of G. Knowles, B. Williams and L. Taylor A corpus of formal British English speech and G. Knowles, A. Wichmann and P. Alderson Working with speech: perspectives on research into the Lancaster/IBM Spoken English Corpus. In: English Language and Linguistics, 1.1, 196-199.
Nelson, G. (1997) Cleft constructions in spoken and written English. Journal of English Linguistics, 25.4, 340-348.
Svartvik, Jan and Alex Chengyu Fang, (1997) Speechmaker. In: To explain the present: Studies in the changing English Language in honour of Matti Rissanen. Terttu Nevalainen and Leena Kahlas-Tarkka (eds.) Helsinki: Société Néophilologique. Mémoires de la Société Néophilologique de Helsinki, LII, 431-450.
Wales, K. (1996) Personal pronouns in presentday English. Cambridge: Cambridge University Press.
Wallis, Sean A. and Gerry Nelson (1997) Syntactic Parsing as a Knowledge Acquisition Problem. In: E. Plaza and R. Benjamins (1997), Knowledge Acquisition, Modeling and Management. Proceedings of the 10th European Workshop, EKAW 1997. Berlin: Springer Verlag, 285300.
Please let us know if you would like us to include any of your publications based on SEU material.
This page last modified 29 April, 2013 by Survey Web Administrator.